Nonlinear model predictive control leveraging quantum-inspired optimization in the three body problem with uncertainty.

The optimization and control of spacecraft trajectories in presence of uncertainties and unmodeled disturbances represents a challenging problem. Future missions would benefit from robust control techniques to compute their trajectory in a complex, nonlinear environment and in presence of stochasticity. In this work the Stochastic Optimal Control problem in the Circular Restricted Three Body Problem is addressed and a method is proposed based on Nonlinear Model Predictive control (NMPC). This method presents the challenge of solving a nonlinear optimization problem on-board. A Quantum-inspired meta-heuristic optimization algorithm is tested in this context: the Diffusion Monte Carlo (DMC) method is based on a numerical techniques used in Quantum Mechanics, adapted to solve optimization problems. The NMPC with DMC is implemented, taking into account an estimate of the computational time required through the introduction of delayed actuation of the control. Two applications are presented featuring transfer trajectories between planar periodic orbits in the vicinity of the Moon; the proposed method proved able to successfully generate trajectories when disturbance accelerations and uncertain initial conditions are present. • The Circular Restricted Three Body Problem with uncertainties is considered. • A NMPC is used to drive the spacecraft to the desired orbit despite the uncertainty. • A Quantum-inspired meta-heuristic optimizer is used in the MPC. • Applications are presented regarding transfers between Lunar periodic orbits. [ABSTRACT FROM AUTHOR]